This paper presents a distributed, linear-time algorithm for localizing sensor network nodes in the presence of noisy range measurements. The algorithm introduces the concept of *robust quadrilaterals* to avoid flip ambiguities that can corrupt localization computations. The localization problem is formulated as a two-dimensional graph realization problem, where the goal is to recover the Euclidean positions of vertices in a planar graph given approximately known edge lengths. The algorithm is implemented on a physical sensor network and its accuracy and performance are empirically assessed. Simulations demonstrate that the algorithm scales well to large networks and handles real-world deployment geometries. The algorithm also supports localization of mobile nodes by recomputing cluster localizations as nodes move. The paper discusses related work, challenges in network localization, and provides a detailed analysis of the algorithm's robustness and computational complexity. Experimental results show that the algorithm performs well relative to the quality of the distance measurements and provides good localization coverage.This paper presents a distributed, linear-time algorithm for localizing sensor network nodes in the presence of noisy range measurements. The algorithm introduces the concept of *robust quadrilaterals* to avoid flip ambiguities that can corrupt localization computations. The localization problem is formulated as a two-dimensional graph realization problem, where the goal is to recover the Euclidean positions of vertices in a planar graph given approximately known edge lengths. The algorithm is implemented on a physical sensor network and its accuracy and performance are empirically assessed. Simulations demonstrate that the algorithm scales well to large networks and handles real-world deployment geometries. The algorithm also supports localization of mobile nodes by recomputing cluster localizations as nodes move. The paper discusses related work, challenges in network localization, and provides a detailed analysis of the algorithm's robustness and computational complexity. Experimental results show that the algorithm performs well relative to the quality of the distance measurements and provides good localization coverage.